Matteo Ferla retweeted
OPIGlets are out in force at #AIChem24, organised by our very own @gmm DPhil students @LucyVost, @arun_raja007 and @isakvals are presenting posters, and Yael is giving a talk about her recent work MolSnapper on Wednesday! Please come and chat to them!
Matteo Ferla retweeted
DPhil student Steph Wills has led efforts to expand catalogue searching via bioisosteric fragment merges, using a graph database approach.
Preprint now available on the bioRxiv: biorxiv.org/content/10.1101/…
Matteo Ferla retweeted
DPhil student @oliverturnbull1 has built pIg-Gen, a language model for paired-chain antibody generation. It creates diverse, natural-like distributions of antibodies w/ similar biophysical properties to therapeutics.
bioRxiv: biorxiv.org/content/10.1101/…
Code: github.com/oxpig/p-IgGen
Matteo Ferla retweeted
OPIG DPhil student @GemmaLGordon led work to build and analyse "PLAbDab-nano: a database of camelid and shark nanobodies from patents and the literature". Just released on bioRxiv and available as an OPIG webapp:
Preprint: biorxiv.org/content/10.1101/…
Webapp: opig.stats.ox.ac.uk/webapps/…
Matteo Ferla retweeted
Generative models for molecular optimization & protein design often rely on data-driven guidance functions for conditional sample generation. Our new #ICML2024 paper presents a simple but effective approach to improve their performance in OOD settings.
arxiv.org/abs/2407.11942
RFdiffusion is sometimes quirky, often infuriating, but most often fun. As a result, I wrote a long blogpost on how to manipulate and score RFdiffusion models in PyRosetta 📚🛠️blopig.com/blog/2024/05/pyro…
Last month I attended RSC Fragments 2024, which was very interesting. Here are my take home concepts: blopig.com/blog/2024/04/rsc-…
1
277
If you are at #Fragments24 do check out poster 26 for in silico fragment elaborations!
It took a while but it's here: the Fragmenstein preprint!
chemrxiv.org/engage/chemrxiv…
A big thank-you to everyone who helped in its creation, especially user feedback!
Before the break I presented at the AViDD forum (asapdiscovery.org/forum/) for which I am very grateful for, but totally forgot to share the link: youtube.com/watch?v=kieDWYkz…
(Title: Advancing Fragment-Based Drug Discovery in the @asap_discovery consortium with Fragmenstein)
Matteo Ferla retweeted
The incredible @asap_discovery team members working with @XChem_DLS collected 14,631 X-ray datasets and deposited 6.7% of all X-ray structures in the PDB for ASAP in 2023 to advance structure-based antiviral discovery--with hundreds more pending release. 🚀

ALT Slide summarizing statistics from the ASAP Structural Biology Core: Collected 14,631 datasets as of 4th December 2023. Rolling average consistently >1,000 samples/month for Q4 2023 Originally predicted ~500 samples/month 601 structures deposited in PDB by ASAP in 2023 (6.7% all X-ray structures) 924 for all of Diamond, 776 for all American synchrotrons >200 pending release
Matteo Ferla retweeted
Here's what I spent the last few years working on. Incredibly proud of the work done by all of the amazing scientists and volunteers who contributed, and optimistic for the future of open science drug discovery for globally equitable and affordable access:
nitter.gq/ScienceMagazine/status…
The COVID Moonshot—an open-science, crowdsourced, and patent-free drug discovery campaign targeting the #SARSCoV2 virus—has yielded a wealth of data that could pave the way for the development of new and better therapeutics.
Learn more in Science: scim.ag/4LB
The @covid_moonshot paper is out in Science: but can it help you as a compchem algo testset? Yes!
All steps (frag to lead) crystallised & assayed, incl. negatives. Diverse methods, 4 series. Don't take my word, have a gander yourself: fragalysis.diamond.ac.uk/vie…
Matteo Ferla retweeted
New & impressive @ScienceMagazine
Here's how you discover a potent new drug vs Covid
212 scientists/15 countries
Open-science, patent free, high-throughput structural biology based design, #AI @covid_moonshot @melissa_boby @FrankvonDelft @Darenfearon
science.org/doi/10.1126/scie…
Matteo Ferla retweeted
PDB released 478 SARS-CoV-2 structures this week, including 466 mpros with inhibitors from the COVID Moonshot project. 966 have been released this year (shown), for 3826 total since the first CV19 structure was released in Feb 2020.
Explore them all at rcsb.org/covid19
For anyone at #shef2023 last week, here are my slides github.com/matteoferla/Fragm… and the interactive table of the follow-up compounds from the Covid Moonshot dataset results (michelanglo.sgc.ox.ac.uk/r/f…)
2
368
Matteo Ferla retweeted
At the @OxfordBRC open day. New t-shirts, new venue, same strawberry 🍓 dna demo
Matteo Ferla retweeted
And now we’re married! @matteoferla
Matteo Ferla retweeted
thanks to everyone who helped with the study which includes @GeCIPteam @GeneDx @ACOffiah @melitairving @RMaroofian @BaylorGenetics @OxfordBRC and @matteoferla (sorry your lovely fig had to move to suppl)
I was curious so I checked what were the most read-about compounds in en-Wikipedia: obviously narcotics dominate the list. However, at 5k views/mo limonene is the most read, which is very peculiar... Any ideas?
(data: github.com/matteoferla/Wikip…)






















